This project proposes to analyze 3 existing sets of data of personal interviews, 1 clinical and 2 community samples, to contrast the emergence, severity and outcomes of alcoholism in these 3 populations. The first is a sample of 393 subjects, initially identified by a hospital discharge diagnosis of alcoholism, who participated in a follow-up study on moderate drinking. The second is 386 household subjects diagnosed by DSM-III criteria as alcoholic in the St. Louis component of the NIMH-funded ECA survey. The third includes 77 alcoholics from an offshoot of the St. Louis ECA survey that assessed the impact of disasters on mental health. Specific aims include comparing emergence of symptoms using both reported age of onset and chronological ranking of symptoms in each sample. Severity of alcoholism, including duration, number of lifetime symptoms, psychiatric comorbidity and number and types of medical complications, will be contrasted across the 3 samples. Outcomes, in terms of return to moderate drinking, remissions lasting 1-3 years and longer than 3 years, and severe alcoholism with medical complications will be compared. The relationship between alcohol consumption and occurrence of specific life events and to development of symptoms of alcoholism will be studied in the clinical sample. Multivariate analytical techniques, including logistic regression and survival analysis, will be employed to test specific hypotheses. Expected benefits of the proposed analyses include identification of a path to alcoholism with points at which preventive intervention strategies would be indicated, differences in caseness between clinical and community samples, predictors of outcomes, and illumination of the relationship between level of alcohol consumption and appearance of specific alcohol symptoms.